<![CDATA[Purpose: Mussaenda frondosa is a rare plant species native to Indonesia’s tropical forests, with limited research focused on its classification and identification, particularly using machine learning. This study aims to develop a classification model for Mussaenda species using a Convolutional Neural Network (CNN) approach to support the advancement of automated plant identification systems. Methodology/approach: The dataset used consists of 650 labeled images, categorized into six primary parts of the plant: leaves, stems, twigs, fruits, flowers, and trees. A CNN model was developed and trained over 200 epochs to classify the images according to these categories. Preprocessing techniques such as resizing, normalization, and data augmentation were applied to enhance model performance. Results/findings: The trained CNN model achieved an accuracy of 80%, demonstrating its ability to classify Mussaenda frondosa components despite the relatively small dataset. Visual inspection of prediction outputs showed consistent identification across several categories, particularly leaves and flowers. Conclusion: The results suggest that CNN can be effectively used to classify rare plant species like Mussaenda frondosa. The model's performance also indicates that even a limited dataset, when properly processed, can yield promising classification results. Limitations: The main limitation of this research is the small dataset size, which may restrict the model's generalizability to broader plant species or more diverse environmental conditions.. Contribution: This study contributes to the field of plant classification by providing a foundation dataset and a validated CNN model for rare tropical species. It opens pathways for further research in biodiversity monitoring and conservation using AI.]]>
